3456
J . Phys. Chem. 1993, 97, 3456-3459
Gas-Pbase Reactions between Hydrocarbons and Metal Oxides: The A I 0 590 to 1380 K
+ CH4 Reaction from
David P. Belyung and Arthur Fontijn' High- Temperature Reaction Kinetics Laboratory, The Isermann Department of Chemical Engineering, Rensselaer Polytechnic Institute, Troy, New York 12180-3590
Paul Marshall' Department of Chemistry, University of North Texas, Denton, Texas 76203-5068 Received: November 16, 1992; In Final Form: February 2, 1993
A recent a b inito study suggested that gaseous metal oxides can directly abstract H atoms from hydrocarbons, but with considerable barriers. To try to confirm the occurrence of such reactions, the title reaction has been studied in a high-temperature fast-flow reactor. The data are well described by the fitting expression k(5901380 K) = 9.6 X 10-39(T/K)7.96exp(2468 K/T) cm3 molecule-' s-I, with 2u precision limits varying with temperature from f 4 % to 112%. The corresponding 2u accuracy limits are about 125%. Comparison of this result to further a b initio and conventional transition-state theory calculations suggests that direct abstraction indeed can occur a t the higher temperatures but that one or more other channels, possibly involving A I 0 insertion into a C-H bond, dominate in the initial attack step. The potential implications of this work for catalytic conversion of methane to higher hydrocarbons are considered.
Introduction In a recent ab initio study Bsrve and Petterssonl predicted that somegaseous metaloxides, LiO, MgO, and A10, can directly abstract an H atom from methane to produce methyl radicals. Such exothermic reactions have not been observed in the gas phase, and their studies suggested these to have considerable barriers, 25-67 kJ mol-'. However, solid metal oxides have been found to be effective catalysts for H abstraction from methane, leading to C2H4 and C2H6 in significant yields.2 a process of considerable commercial promise. The basis for this catalytic reaction has been strongly suggested to be the surface reaction of the 0-ion to remove a hydrogen atom, a process which appears to proceed to a significant degree only above about 1000 K.3 Theoretical studies of cluster models of the MOO, oxide structure have supported this hypothe~is.~If the occurrence of the gasphase processes could be experimentally confirmed, it would not only establish a new class of reactions but also could lead to a better understanding of the heterogeneous catalytic processes. We therefore decided to investigate one of these suggested reactions: A10
+ CH,
-
AlOH
+ CH,
(1) for which AHo = -27 kJ mol-l.s.6 While this reaction has a higher predicted barrier than those of the other oxides,' we have had extensiveexperiencewith thestudy of the gas-phase reactions of A10 at temperatures up to 1600 K in HTFFRs (hightemperature fast-flow reactor^)^ and therefore selected this oxide.
Technique The HTFFR facility used in this work and the data handling procedures have been previously described.' Briefly, a vertical mullite (McDanel MV-30) reaction tube (60 cm long, 2.2 cm i.d.) is heated to the desired temperature by columns of resistively heated S i c rods inside an insulated,water-cooled vacuum housing. A1 vapor is produced by resistively heating an Al-wetted W coil in a flow of Ar carrier gas. A trace of oxidizer, 0.5% N 2 0or 5% 02 in Ar, at -4 X of the main Ar flow, is passed through a Pt side tube to a location just downstream from thevapor source to rapidly convert A1 to A10.8 '0 Further downstream CH4/Ar mixtures at flow rates of 4 8 % of the main Ar flow rate are 0022-3654/93/2091-3456$04.00/0
introduced through a movable inlet. Rate coefficient measurements are made under pseudo-first-order conditions, [A101 0), then dissociation of the excited adduct back to A10 CH4 would be less competitive with collisional stabilization and the predicted rate coefficients would be even closer to the high-pressure limit. The observed pressure independence is also consistent with a mechanism where the adduct fragments to new products before stabilization. Further QRRKcalculations suggest that formation of any adduct, which is sufficiently thermodynamically stable to be significant in our experiments, will have a rate coefficient that is in the falloff region or close to the high-pressure limit.
+
+
+
-
+
Conclusions These experiments are in accord with the suggestion by B0rve and Petterssonl that gas-phase reactions between A10 and CHI occur and may provide insight in the heterogeneous catalytic conversion of methane to C Z hydrocarbons. However, use of their calculated barrier suggests that the branching ratio for direct abstraction is small and therefore that other reaction channels are important. Our calculations demonstrate that insertion leading to CH3AlOH is a reasonable candidate for the initial step, although other channels are not excluded. The insertion
products (Figure 2b-d) might model the products of dissociative adsorption of CHI onto an alumina catalyst, where their energies are likely to differ from gas-phase values. More detailed experimental and theoretical investigations of the present reaction system, as well as of other metal oxide-hydrocarbon reactions, are needed to obtain an understanding of this new class of gasphase reactions and their potential importance for methane conversion.
Acknowledgment. The experimental work was performed at Rensselaer under AFOSR Grants F4962O-92-5-0 172and F49620. 92-5-0346, and the theoretical study was done a t the University of North Texas with support from the Robert A. Welch Foundation (Grant B-1174) and the U N T Organized Research Fund. References and Notes ( I ) Bsrve, K. J.; Pettersson, L. G . M. J. Phys. Chem. 1991, 95, 3214. (2) Feng, Y.; Niiranen, J.; Gutman, D. J . Phys. Chem. 1991,95,6564. (3) Hutchings. G . J.; Woodhouse, J. R.; Scurrell, M. S. J . Chem. Soc., Faraday Trans. I 1989.85. 2507. (4) Mehandru, S.P.; Anderson, A. B.; Brazdil, J. F.; Grasselli, R. K. J . Phys. Chem. 1987, 91, 2930. ( 5 ) Pedley, J. B.; Naylor, R. D.; Kirby, S. P. Thermochemical Data of Organic Compounds; Chapman and Hall: New York, 1986; p 89. (6) Chase,M.W.,Jr.;Davies,C.A.;Downey,Jr.;Frurip,D.J.;McDonald, R. A.; Syverup, A. N. JANAF Thermochemical Tables;J. Phys. Chem. ReJ Data 1985, 14 (Suppl. No. I). (7) Slavejkov, A. G.; Stanton, C. T.; Fontijn, A. J. Phys. Chem. 1990. 94, 3347 and references therein. (8) Parnis, J. M.; Mitchell, S. A.; Kanigan,T. S.; Hackett, P. A. J . Phys. Chem. 1989, 93, 8045. (9) Fontijn, A.; Futerko, P. M.In Gas-Phase Metal Reactions; Fontijn, A., Ed.; North-Holland: Amsterdam, 1992; Chapter 6. ( I O ) Garland, N. L.; Nelson, H. H. Chem. Phys. Lett. 1992, 191. 269. ( I I ) Fontijn, A.; Felder, W. In Reactive Intermediates in the Gas Phase. Generation and Monitoring, Setser, D. W., Ed.; Academic Press: New York, 1979; Chapter 2. (12) Pearse, R. W. B.; Gaydon, A. G. The Identijication of Molecular Spectra; Chapman and Hall: London, 1976; p 41. (13) Irvin, J. A.; Quickenden, T. I. J . Chem. Educ. 1983, 60, 71 I . (14) Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W. T. Numerical Recipes; Cambridge University: Cambridge, 1986; Chapter 14. ( I 5) Benson, S.W. ThermochemicalKinerics, 2nd ed.; John Wiley: New York, 1976; Chapter 4. (16) Wagener, R.; Wagner, H. Gg. Ber. Bunsen-Ges. Phys. Chem. 1990, 94, 1096. (17) Chu, J. 0.;Beach, D. B.; Jasinski, J. M. J . Phys. Chem. 1987, 91, 5340. (18) Becerra, R.; Frey, H. M.; Mason, B. P.;Walsh, R.; Gordon, M. S. J . Am. Chem. SOC.1992, 114. 2752. (19) Wentworth, W. E. J . Chem. Educ. 1965.42, 96. (20) Fontijn, A.; Felder, W. J . Phys. Chem. 1979, 83, 24. (21) Frisch. M. J.; Head-Gordon, M.; Trucks, G . W.; Foreman, J. B.; Schlegel, H. B.; Raghavachari, K.; Robb, M. A.; Binkley. J. S.;Gonzalez, C.; Defrees, D. J.; Fox, D. J.; Whiteside, R. A.; Seeger, R.; Melius, C.F.; Baker, J.; Martin, R. L.; Kahn, L. R.; Stewart, J. J. P.; Topiol, S.;Pople, J. A. Gaussian 90; Gaussian: Pittsburgh, PA, 1990. (22) Yu, H.; Goddard, J. D. Can. J . Chem. 1990, 68,663. (23) Jeong, G . H.; Klabunde, K. J. J . Am. Chem. Soc. 1986,108,7103. (24) Jordan, K. D.; Kurtz, H. A. In Metal Bonding and Interactions in High TemperatureSystems; ACSSymposiumSeries I79;Gole.J. L.,Stwalley, W. C., Eds.; American Chemical Society: Washington, DC. 1982; Chapter 26. (25) Oblath, S. B.; Cole, J. L. J . Chem. Phys. 1979, 70, 581; Combust. Flame 1980, 37, 293. (26) Unscaled MP2/3-21G(*) frequencies Y for CHIAIOH are 21, 220, 387.520,618,662,759,935, 1322, 1549, 1559,3054,3130, 3155.and 3674 cm '. Y for CHrAI(0)H are44,192,4I I , 581,655,744,781,908,1356,1555, 1557, 2016, 3057, 3125, and 3150 cm I; Y for CHiOAlH are 87, 151. 242, 630,726, 1158, 1203, 1275, 1557, 1607, 1613, 1910, 3061, 3126, and 3142 cm I. The C H I groups in these species are essentially free rotors. (27) Difficulties in describing A I 0 with a single reference wave function make these estimates uncertain, although the energy difference between the three adducts is unaffected. See e.&: Marshall, P.; OConnor, P. B.; Chan. W. T.; Kristof, P. V.; Goddard. J. D. In Gas-Phase Metal Reactions; Fontijn. A,, Ed.; North-Holland: Amsterdam, 1992; Chapter 8. (28) (a) Dean, A. M. J . Phys. Chem. 1985,89,4600. (b) Westmoreland. P. R. QRRK EGA Program, 1988.
